WO2004043079A1

WO2004043079A1 - Three-dimensional video processing method and three-dimensional video display

Info

Publication number: WO2004043079A1
Application number: PCT/JP2003/014166
Authority: WO
Inventors: Ken Mashitani; Goro Hamagishi
Original assignee: Sanyo Electric Co., Ltd.
Priority date: 2002-11-07
Filing date: 2003-11-06
Publication date: 2004-05-21
Also published as: KR20050084950A; EP1581012A1; JPWO2004043079A1; US20060125916A1; CN1708995A; KR100785982B1

Abstract

A method for processing a video by extracting dots of three colors as a pixel unit for each of view point videos. The extracted three-color dots are made a pixel group. A pixel group is composed of 21 dots surrounded by the heavy line. An opening (1) corresponds to each pixel group. From a view position, three-color dots (4, 11, R), (4, 11, G), (4, 11, B) constituting a pixel at the upper left of viewpoint 4 can be seen simultaneously . When the viewpoint moves, a pixel of different viewpoint such as composed of (3, 11, R), (3, 11, G), (3, 11, B) of the same pixel coordinate 11 can be seen. The pixels of one pixel group are arranged so that the aspect ratio of the display pitch of the pixel group may be most approximate to 1:1. Thus, even if the number of viewpoints is large, the degradation of the horizontal resolution can be mitigated, and hence the quality of image can be improved.

Description

Description 3D image processing method and 3D image display device

The present invention relates to a stereoscopic video display device and a stereoscopic video processing method capable of performing stereoscopic vision without requiring special glasses. Background art

Conventionally, a parallax barrier method, a lenticular lens method, and the like have been known as methods for achieving stereoscopic image display without the need for special glasses, but these methods use an eye having binocular parallax. For example, the image for the left eye and the image for the left eye are alternately displayed on the screen, for example, in the form of a vertical stripe, and the displayed images are separated by a single lens such as a paralux barrier lens and guided to the right and left eyes of the observer. This is to perform stereoscopic vision.

FIG. 11 is an explanatory diagram exemplifying the principle of a stereoscopic image display device of a four-lens type stereoscopic vision system. Image 1, image 2, image 3, and image 2 having a binocular disparity in the horizontal direction of screen 11 are arranged at a predetermined pitch, and a unit image group of “image 2 image 2 image 3 image 2” repeatedly exists. I have. The apertures 12a of the image separation means 12 are provided corresponding to each unit image group, and each of the unit image groups "image 1 image 2 image 3 image 2" is separated and given to the observer.

FIG. 12 shows the arrangement of pixels and display pixel data in the seven-view stereoscopic system. In the figure, a group of 2 "I pixels (dots) surrounded by a bold line constitutes one pixel group, which corresponds to three pinholes (openings). The upper left picture element (pixel) Dots of {1; 11; R}, {1; 11; G} {1; 11; B} are simultaneously observed. Here, {i; jk; C} represent {viewpoint; pixel coordinates; color}, respectively. Then, according to the movement of the observation position, pixels having different viewpoints such as {2; 11; G}, {2; 11; B}, {2; 11; Will be observed.

In the stereoscopic video processing method using the pixel arrangement and display pixel data shown in Fig. 12 (a), only the number of pixels in the horizontal direction is reduced to 1/7. This means that the horizontal pitch of the pixels is increased by a factor of seven, resulting in a ratio of horizontal to vertical pitch of seven.

If this concept is applied to a seven-lens stereoscopic image display device using video separation means (not shown) in which pinholes are arranged diagonally, the result is as shown in Fig. 12 (b). The pixel group is the same, but the arrangement of viewpoint numbers in the group changes. Here, the number of pixels in the horizontal direction is similarly degraded. If there is parallax only in the horizontal direction, the number of pixels usually deteriorates only in the horizontal direction. Japanese Patent No. 3096613 is known as a stereoscopic video display device in which a plurality of pinholes are arranged diagonally.

In a general video display device, adjacent three-color pixels (dots) of red, green, and blue constitute one picture element (pixel). In the parallax barrier method and the stereoscopic image display device disclosed in Patent Document 1, as shown in FIG. 12, dots that originally constitute the same pixel display images from different viewpoints. Also, when the number of viewpoints increases, dissatisfaction occurs such as inevitable reduction in the horizontal resolution in the combination of dots constituting the pixels of each viewpoint. Also, there was no proposal for a video processing method suitable for a stereoscopic video display device. Disclosure of the invention

The present invention has been made in view of the above circumstances, and has as its object to provide an improved stereoscopic video processing method and stereoscopic video display device.

In order to solve the above-mentioned problem, a stereoscopic video processing method of the present invention is a stereoscopic video processing method for extracting a plurality of pixels serving as picture element units for each viewpoint video from the plurality of viewpoint videos. A group of data of a plurality of pixels, which is a pixel unit extracted from the image data, is defined as a pixel group, and the aspect ratio of the display pitch of the pixel group on the screen of the stereoscopic video display device is closest to 1: 1. The arrangement of picture element units in the picture element group is set.

With the above configuration, the arrangement of the picture element units in the picture element group is set so that the aspect ratio of the display pitch of the picture element group on the screen of the stereoscopic video display device is closest to 1: 1. Therefore, it is preferable that the pixels constituting the picture element of each viewpoint are close to each other, for example. Even if the number of viewpoints increases, the decrease in the horizontal resolution can be eased, so that an improvement in image quality can be expected.

Also, the stereoscopic video processing method of the present invention is a stereoscopic video processing method for extracting a plurality of pixels that are picture element units for each viewpoint video from a plurality of viewpoint videos, and includes a picture element unit extracted from each viewpoint video. A group of data of a plurality of pixels is defined as a pixel group, and the aspect ratio of the display pitch of the pixel group on the screen of the stereoscopic video display device is in a range of 1: 2 to 2: 1. The arrangement of picture element units in the picture element group is set.

In such a configuration, the arrangement of the picture element units in the picture element group is such that the aspect ratio of the display pitch of the picture element group on the screen of the stereoscopic video display device is in the range of 1: 2 to 2: 1. Is set, it is preferable that the pixels constituting the picture element of each viewpoint come closer to each other, and the decrease in the horizontal resolution can be reduced even when the number of viewpoints increases. The data of a plurality of pixels as a picture element unit extracted from each viewpoint video may be obliquely arranged on a bit map. Further, data of a plurality of pixels, which are pixel units extracted from each viewpoint video, may be supplied so as to be obliquely arranged on the screen of the stereoscopic video display device. Further, data of a plurality of pixels, which is a picture element unit extracted from each viewpoint video, may be converted into a video signal so as to be obliquely arranged on the screen of the stereoscopic video display device and supplied.

The number of display picture elements is horizontal M x vertical N, the number of viewpoints is assumed, the number of pixels constituting one picture element is k, and the number of picture elements of each viewpoint video is horizontal k M / L x vertical N / k. Necessary pixel data may be extracted from each viewpoint video for each video region corresponding to the video. According to this, the coordinates of the pixels in each viewpoint video are not inherited in the stereoscopic video, but since there are no pixels discarded in each viewpoint video, wasteful video generation can be eliminated. In addition, the images acquired by the image acquisition system are processed such that the number of picture elements of each viewpoint video is horizontal k M / L x vertical N / k and the image aspect ratio matches the aspect ratio of the display image, You may acquire each viewpoint video. According to this, distortion of the display image can be prevented. Also, the image aspect ratio of the image acquisition system may be made to match the aspect ratio of the display image to acquire each viewpoint video. According to this, the distortion of the display image can be prevented.

The number of picture elements in the display is horizontal MX vertical N, the number of viewpoints is assumed, the number of pixels constituting one picture element is k, and the number of picture elements of each viewpoint video is obtained as horizontal k M / LX vertical NZ k The viewpoint video may be enlarged to horizontal MX and vertical N, and necessary pixel data may be extracted and generated from each viewpoint video for each corresponding video area. According to this, the distortion of the display image can be prevented. In addition, this processing method requires more memory for compositing, but the load on the image acquisition system is smaller than when acquiring the display image (composite image) size from the beginning. Each viewpoint video may be one pixel or several pixels larger on the left and right, and data extracted from the enlarged pixels may be used in non-data portions that occur on the left and right of the screen. Alternatively, black data may be used in non-data portions that occur on the left and right sides of the screen. Alternatively, copy data of adjacent pixels having the same viewpoint may be used.

Also, a stereoscopic video image having a vertical parallax may be generated.

Further, the stereoscopic video display device of the present invention is a stereoscopic video display device comprising: a screen on which a video is displayed; and a separating unit for separating a position where a pixel of each viewpoint video can be observed. When the image obtained by the image processing method is displayed on the screen, the aspect ratio of the screen pixel pitch is set so that the aspect ratio of the pitch of the displayed picture element group on the screen is 1: 1 to approximately 1: 1. It is characterized by having. If the number of viewpoints is L and the number of pixels constituting one picture element is k, the display pixel pitch is set to k (horizontal): L (vertical), and the aspect ratio of the display picture element group is horizontal: vertical = It may be configured to be 1: 1.

In the above configuration, the pixel row for red, the pixel row for green, and the pixel row for blue may be arranged sequentially in the vertical direction. According to this, the arrangement of the colors of the pixels constituting the picture element is changed. The image quality of the screen edge is improved. Also, the stereoscopic video display device of the present invention is a stereoscopic video display device including a screen on which a video is displayed and a separation unit for separating a position where a pixel of each visual point video can be observed, wherein The aspect ratio of the screen pixel pitch is set to k L to 1 to approximately k L to 1, and the pixel data of each viewpoint video is supplied in the horizontal direction, and the video is displayed. The display is characterized in that the aspect ratio of the pitch of the picture element groups is 1: 1 to approximately 1: 1. In such a configuration, the number of viewpoint videos It is preferable that the same number of pixels of the same color are arranged consecutively. According to this, the arrangement of the colors of the pixels constituting the picture element matches, and the image quality of the screen edge is improved. BRIEF DESCRIPTION OF THE FIGURES

FIG. 1 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on a pixel. FIG. 2 is an explanatory diagram showing the color arrangement of the screen in FIG. FIG. 3 is an explanatory diagram showing a process of combining a display image from a plurality of original images. FIG. 4 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a process of combining a display image from a plurality of original images. FIG. 5 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on a pixel. FIG. 6 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on a pixel. FIG. 7 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen ±, a size of a pixel area, and an image displayed on a pixel. FIG. 8 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on the pixels. FIG. 9 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on the pixel. FIG. 10 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on a pixel. FIG. 11 is an explanatory diagram showing a basic configuration of a multi-view stereoscopic image display device. FIGS. 12 (a) and 12 (b) are diagrams each showing a conventional example, and are explanatory diagrams showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on a pixel. FIG. 13 is a diagram showing an embodiment of the present invention. FIG. 4 is an explanatory diagram showing a process of combining a display image from a plurality of original images. FIG. 14 is a diagram showing the embodiment of the present invention, and is an explanatory diagram showing a process of synthesizing a display image from a plurality of original images. FIG. 15 is a diagram showing the embodiment of the present invention, and is an explanatory diagram showing a process of synthesizing a display image from a plurality of original images. FIG. 16 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on the pixel. FIG. 17 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on the pixels. FIG. 18 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on the pixel. FIG. 19 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on the pixels. FIG. 20 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on the pixel. FIG. 21 is a diagram showing an embodiment of the present invention, and is an explanatory diagram showing a color arrangement of pixels on a screen, a size of a pixel area, and an image displayed on a pixel. FIG. 22 is an explanatory diagram showing the processing of FIG. BEST MODE FOR CARRYING OUT THE INVENTION

Hereinafter, a stereoscopic video processing method and a stereoscopic video display device according to an embodiment of the present invention will be described with reference to FIGS. 1 to 9 and FIGS. Note that the entire configuration of the stereoscopic video display device can adopt the configuration of FIG. 11 described in the conventional section, and description of the entire configuration is omitted to avoid redundancy due to duplication of description.

Figure 1 shows the color arrangement of pixels (dots) on the screen of the stereoscopic image display device (R, G, and B columns), the size of the pixel area, and the images displayed on the pixels. Here, the number of each viewpoint video is assumed to be 7 (seven-eye system), and red, green, and blue (RG B), which are the picture elements (pixels) extracted from each viewpoint video, for each image region corresponding to each viewpoint video The pixel data is given to the obliquely arranged dots, and the image is separated, for example, by the aperture 1 shown by the dotted line in the figure. 21 dot groups surrounded by a bold line constitute one pixel group, and the aperture 1 corresponds to this pixel group. From a certain observation position, the upper left pixel of the viewpoint 1 is constituted {1; Dots of three colors of 11; R}, {1; 11; G}, and {1; 11; B} are observed at the same time. Here, {i; jk; C} represent {viewpoint; pixel coordinates; color}, respectively. Then, according to the movement of the observation position, different viewpoints such as {2; 11; G}, {2; 11; B} _% {2; 11; R} ■■■ at the same pixel coordinates 11 Observe the pixels.

A screen (display) that optimally displays a planar image is used. Here, as an example, it is assumed that a liquid crystal panel having 3840 horizontal pixels × 2400 vertical pixels is used. Each pixel is composed of a combination of dots of three colors, red, green and blue. As shown in Fig. 2, dots of the same color are arranged vertically, and the aspect ratio of this dot pitch Is three-to-one, and the horizontal and vertical pitches of pixels in a two-dimensional image display are equal. In other words, the aspect ratio of the pixel pitch is 1: 1, which is the desired value for flat image display.

Here, the number of horizontal and vertical pixel groups in the stereoscopic video processing method shown in Fig. 12 in the conventional section exists on a screen with 3840 horizontal pixels × 2400 vertical pixels as shown below. Will be. The values in parentheses are example values.

Number of viewpoints L (7)

Number of pixels in display area Horizontal MX Vertical N (3 840 x 2400) Horizontal number of pixel groups M / L (= 548) Vertical number of pixel groups N (2400)

On the other hand, according to the stereoscopic video processing method of this embodiment in which a pixel group is selected as shown in FIG. 1, deterioration of the number of pixels can be dispersed in the vertical direction. Here, one dot is composed by combining three dots arranged diagonally, so the vertical pitch of the pixel group is tripled and the horizontal pitch is 7/3 times. The ratio of the horizontal pitch to the vertical pitch is 7/9, which is closer to 1 which is the value of the ratio of the horizontal pitch to the vertical pitch of pixels in the original flat image display.

This can be generally expressed as follows. The values in the brackets are the values of the embodiment. Number of viewpoints L (7)

Number of pixels in display area Horizontal MX Vertical N (3840 x 2400) Number of dots constituting one pixel k (3)

Horizontal number of pixel groups Mx k / L (= 1 646)

Vertical number of pixel groups N / k (800)

As described above, the arrangement of the display pixels is set so that the aspect ratio of the pitch of the display pixel group on the screen of the stereoscopic image display device is closest to 1: 1. Therefore, the image quality can be expected to be improved because the dots constituting the image are close to each other, and the resolution in the horizontal direction can be reduced even when the number of viewpoints increases.

Here, a process of combining the display image from the L original images will be considered. First, as shown in Fig. 3, for each pixel number MXN (3840 x 2400) of the display image, the number of pixels of each original image (camera captured image, etc.) is also set to MXN (3840 x 2400). There is a method of selecting only dots and synthesizing. In this method, the coordinates of the dots are correctly inherited, but the generation of images is wasted because some dots are discarded. Dots whose pixel coordinates are "1" are discarded. Such a standing A three-dimensional image processing method improved from the body image processing method will be described with reference to FIG.

In the method shown in Fig. 4, the number of pixels of the original image (camera captured image, etc.) is set to k MZL x NZk (1646 X 800), and the dots are synthesized while appropriately rearranging the dots. In this method, the coordinates of the dots are not correctly inherited, but there is no waste of image generation because no dots are discarded. If the image aspect ratio of the image acquisition system is M: N, no image distortion occurs. This is the case where the aspect ratio of the pixel pitch is 1: 1. More generally, if the image aspect ratio of the image acquisition system matches the aspect ratio of the display image, no image distortion occurs. . The image aspect ratio of the image acquisition system is, for example, the aspect ratio of the CCD of a live-action camera or the image aspect ratio at the time of rendering in computer graphics.

Dots with no information are generated on the left and right sides of the screen. (1) Make each viewpoint video one dot or several dots larger than the required dot to the left and right, and place the data extracted from the enlarged dot in non-data locations that will occur on the left and right of the screen. For example, assuming that the pixel coordinates of the enlarged dot are “10”, the non-data portion at the upper left of the composite image includes {7; 10; R}, {7; 10; G}, { 6; 10; R}. (2) Place black data in non-data areas that may occur on the left and right of the screen (non-lighting, light opaque). (3) Place the copy data of the adjacent dots of the same viewpoint. In the example of the figure, {7; 11; R}, {7; 11; G}, {6; 11; R} are arranged.

Figure 5 shows a stereoscopic video display device with a dot pitch aspect ratio that is not 3: 1. When the aspect ratio of the dot pitch is 3: 1, the ratio of the horizontal pitch to the vertical pitch of the display pixel group is 7/9 by the above-described stereoscopic image processing method. In contrast, in Fig. 5, the dot pitch The aspect ratio is 7: 3. As a result, the value P h / PV of the ratio of the horizontal pitch to the vertical pitch of the display pixel group becomes equal to the desired value “1” in the flat image display. ? F) By setting it to “1”, the aspect ratio of the pixel pitch at the time of imaging can be set to 1: 1 and existing equipment and computer programs can be used as they are. FIG. 6 shows a stereoscopic video display device having a color arrangement different from the color arrangement shown in FIG. In the stereoscopic image display device shown in Fig. 1, the red, green, and blue columns are arranged sequentially in the horizontal direction, while the red, green, and blue lines are arranged sequentially in the vertical direction. With such a configuration, the arrangement order of the colors of the dots constituting the pixels is the same, so that the image quality for displaying the edges is improved.

FIG. 7 shows a seven-lens stereoscopic video display device of another embodiment. In this stereoscopic video display device, the aspect ratio of the dot pitch of the screen is set to 21: 1 (kL: 1). Then, one pixel group is composed of 21 dots arranged in the horizontal direction. This allows the value of the ratio of the horizontal pitch to the vertical pitch of the pixel group to be the desired value in the flat image display.

It is equal to "1". Note that when such a dot pitch is adopted, the method of selecting a pixel group changes. The aspect ratio of the dot pitch changes depending on the number of viewpoints.

FIG. 8 shows a seven-eye stereoscopic image display device similar to that of FIG. 7, but the dot arrangement is different. As shown in the figure, seven dots from the left are red, the next seven dots are green, and the remaining seven dots are blue. As a result, the order of the colors of the dots constituting the pixels matches, and the image quality for displaying the edges is improved. In FIG. 9, the pixel arrangement in the display pixel group shown in FIG. 9A is different from the pixel arrangement in the display pixel group shown in FIG. In each case, the diagonal dot system (oblique) Barrier method). The stereoscopic image processing method shown here takes into account the pixel pitch (dot pitch) of the binocular and oblique dot type stereoscopic image display device, and determines the pitch of the display pixel group (see the thick line in the figure) on the screen. The arrangement of pixels in the display pixel group can be switched to either Fig. 9 (a) or Fig. 9 (b) so that the aspect ratio is closest to 1: 1. For example, an image processing device capable of generating the images shown in FIGS. 9 (a) and 9 (b) can be used, and a three-dimensional image display device connected to the image processing device can be a liquid crystal display panel or a plasma display device. Even if the pixel pitches are different from each other when purchasing a display, the aspect ratio of the pitch of the display pixel group can be reduced to 1: 1 by generating the image shown in either Fig. 9 (a) or Fig. 9 (b). Can be approached.

FIG. 10 shows a configuration example in a case where parallax is also provided in the vertical direction. The horizontal and vertical dot pitches are made equal so that the distance between the horizontal and vertical viewpoints is the same. In order to increase the number of eyes in the horizontal direction, the dots are arranged in the same color in the vertical direction. With this configuration, the ratio of the horizontal pitch to the vertical pitch of the pixel group becomes 1 in the case of the horizontal 6-view system and the vertical binocular system. The order of the colors of the dots that make up the pixels will also match. Note that the distance between the horizontal and vertical viewpoints does not necessarily have to be the same.

FIG. 13 shows a modified example of the method shown in FIG. 4, which shows no image distortion. In this example, the number of picture elements to be displayed is horizontal M (3840) X vertical N (240 0), the number of viewpoints is L (7), the number of pixels that constitute one picture element is k (3), A method of extracting necessary pixel data from each viewpoint image for each image region corresponding to each viewpoint image by setting the number of picture elements of the image to horizontal k M / L (1 646) X vertical N / k (800), and The aspect ratio of the image acquired by the image acquisition system is calculated as the aspect ratio of the displayed image (2400 (vertical): 3840 (horizontal)). They are matched and each viewpoint video is acquired. In the above case, the aspect ratio of the pixels of each camera is not 1: 1, but horizontal: vertical = 7: 9.

As described above, it is sufficient that the image acquisition system can obtain an appropriate number of pixels and the image aspect ratio, but if it cannot be obtained, image distortion can be prevented by performing expansion processing on each viewpoint video. . Since the pixels of a normal camera are horizontal: vertical = 1: 1 (square), it is better to perform expansion and contraction processing.

FIG. 14 shows an example of the expansion / contraction processing. In this example, the number of picture elements in the display is horizontal M (3840) X vertical N (2400), the number of viewpoints is L (7), and the number of pixels that constitute one picture element is k (3). Let the number of picture elements in the video be horizontal k MZL (1 646) X vertical N / k (800). Then, the image acquired by the image acquisition system (the number of pixels 1 646 (horizontal) X 800 (vertical), the width-to-vertical ratio of the image 1 646 (horizontal) X 800 (vertical)) is converted to the horizontal and vertical Processing is performed so that the ratio matches the ratio of the width and height of the displayed image (3840 (width): 2400 (length)), and each viewpoint video is acquired.

The above processing will be further described with reference to FIG. In FIG. 22, the number of pixels of the camera (the image acquired by the image acquisition system) is 1 024 (horizontal) X 768 (vertical), and the number of viewpoints is 8. If each camera image is multiplied by 98 horizontally, an image of 1 152 (horizontal) X 768 (vertical) is obtained. This is multiplied by 1 to obtain an image of 1500 (horizontal) x 1000 (vertical). When a display image is generated from these eight images, the number of picture elements in the display image becomes horizontal (M = 4000) (1500 x 8/3 = 4000) and vertical (N = 3000) (1000 X 3 ). The screen of the display device is 4 (horizontal): 3 (vertical), and the pixel pitch is 1: 1.

Fig. 15 is an improvement of the processing example of Fig. 3, in which the camera image is converted to the size of the display image (synthesized image) and then synthesized by thinning. In this example, the number of picture elements in the display is horizontal M (3840) X vertical N (2400), Let the number of viewpoints be L (7), the number of pixels that constitute one picture element be k (3), and obtain each viewpoint video (horizontal k M / L (1 6 4 6) X vertical N / k (800 )) Is expanded to horizontal M (3840) X vertical N (2400), and the necessary pixel data is extracted and generated from each viewpoint video for each corresponding video area. With this processing method, the memory used for compositing is increased, but the burden on the image acquisition system (camera and computer graphics processing) is limited by acquiring the display image (composite image) size from the beginning as shown in Fig. 3. Less than compared. Figure 16 shows an example of the display image layout when the screen (color filter) is a horizontal stripe. The arrangement of the display image is a vertical stripe arrangement, and R, G, and B pixels for each viewpoint are formed in the vertical direction. The aspect ratio of the pitch of each pixel is 1 (horizontal): 3 (vertical), and the aspect ratio of the pixel group is horizontal: vertical = 7: 9.

Figure 17 shows an example of the display image layout when the screen (color filter) is a horizontal stripe. The arrangement of the display image is a vertical stripe arrangement, and R, G, and B pixels for each viewpoint are formed in the vertical direction. The aspect ratio of the pitch of each pixel is 3 (horizontal): 7 (vertical), and the aspect ratio of the pixel group is horizontal: vertical = 1: 1. In general terms, the pitch of each pixel is k (horizontal): L (vertical), and the aspect ratio of the pixel group is horizontal: vertical = 1: 1. Note that L is the number of viewpoints and k is the number of dots that make up one pixel.

Figure 18 shows an example of the display image arrangement when the screen (color filter) is diagonal. The arrangement of the display image is a vertical stripe arrangement, and R, G, and B pixels for each viewpoint are formed in the vertical direction. The aspect ratio of the pitch of each pixel is 1 (horizontal): 3 (vertical), and the aspect ratio of the pixel group is horizontal: vertical = 7: 9.

Figure 19 shows the display screen when the screen (color filter) is diagonal. 3 shows an example of image arrangement. The arrangement of the display image is a vertical stripe arrangement, and R, G, and B pixels for each viewpoint are formed in the vertical direction. The aspect ratio of the pitch of each pixel is 3 (horizontal): 7 (vertical), and the aspect ratio of the pixel group is horizontal: vertical = 1: 1. In general terms, the pitch of each pixel is k (horizontal): L (vertical), and the aspect ratio of the pixel group is horizontal: vertical = 1: 1. Note that L is the number of viewpoints and k is the number of dots that make up one pixel.

FIG. 20 shows an example of the arrangement of display images when the screen (color filter) is diagonal. The display image is arranged obliquely, and R, G, and B pixels for each viewpoint are formed in an oblique direction. The aspect ratio of the pitch of each pixel is 1 (horizontal): 3 (vertical), and the aspect ratio of the pixel group is horizontal: vertical = 7: 9.

Figure 21 shows an example of the display image layout when the screen (color filter) is diagonal. The display image is arranged obliquely, and R, G, and B pixels for each viewpoint are formed in an oblique direction. The aspect ratio of the pitch of each pixel is 3 (horizontal): 7 (vertical), and the aspect ratio of the pixel group is horizontal: vertical = 1: 1. In general terms, the pitch of each pixel is k (horizontal): square (vertical), and the aspect ratio of the pixel group is horizontal: vertical = 1: 1. Here, L is the number of viewpoints, and k is the number of dots forming one pixel.

Note that the image separation element is not limited to an opening such as a pinhole, and a lens element may be used. In addition, a configuration in which an image separation unit is disposed on the light source side may be adopted. Also, it is assumed that the pixels that constitute one picture element (pixel) are RGB (K = 3). However, if the pixels that constitute one picture element are R GG 処理, the processing may be performed with Κ = 4.

As described above, according to the present invention, the image forming the picture element of each viewpoint is This is preferable because elements are close to each other, and even when the number of viewpoints is large, a decrease in the horizontal resolution can be reduced, so that various effects such as an improvement in image quality can be expected.

Claims

1. A stereoscopic video processing method for extracting a plurality of pixels as a picture element unit for each viewpoint video from a plurality of viewpoint videos. As a group, the aspect ratio of the display pitch of the picture element group on the screen of the stereoscopic video display device is 1: 1 blue.

A three-dimensional image processing method, wherein the arrangement of the picture element units in the picture element group is set so as to be close to each other.

of

2. A stereoscopic video processing method for extracting a plurality of pixels, which are picture element units for each viewpoint video, from a plurality of viewpoint videos.

A group of data of a plurality of pixels is defined as a picture element group, and the aspect ratio of the display pitch of the picture element group on the screen of the stereoscopic video display device is in the range of 1: 2 to 2: 1. A stereoscopic video processing method characterized by setting the arrangement of picture element units in picture element groups.

3. The stereoscopic video processing method according to claim 1 or 2, wherein data of a plurality of pixels, which are pixel units extracted from each viewpoint video, are obliquely arranged on a bit map. Method.

4. In the stereoscopic video processing method according to claim 1 or 2, data of a plurality of pixels, which are pixel units extracted from each viewpoint video, are supplied so as to be obliquely arranged on a screen of the stereoscopic video display device. Stereoscopic video processing method characterized by performing

5. In the stereoscopic video processing method according to claim 1 or 2, the data of a plurality of pixels, which are picture element units extracted from each viewpoint video, are arranged obliquely on the screen of the stereoscopic video display device. A stereoscopic video processing method characterized by being supplied as a signal.

6. The stereoscopic video processing method according to any one of claims 1 to 5, The number of picture elements on the display is horizontal MX and vertical N, the number of viewpoints is L, the number of pixels constituting one picture element is k, and the number of picture elements of each viewpoint video is horizontal k M / LX vertical N / k. A stereoscopic video processing method characterized by extracting necessary pixel data from each viewpoint video for each video region corresponding to the viewpoint video.

7. The stereoscopic video processing method according to claim 6, wherein the number of picture elements in each viewpoint video is horizontal k M / L x vertical N / k, and the image aspect ratio matches the aspect ratio of the display image. A stereoscopic video processing method characterized by processing images acquired by an image acquisition system and acquiring each viewpoint video.

8. The stereoscopic video processing method according to claim 6, wherein an image aspect ratio of an image acquisition system is made to coincide with an aspect ratio of a display image to acquire each viewpoint video.

9. In the stereoscopic video processing method according to any one of claims 1 to 5, the number of picture elements to be displayed is horizontal MX vertical N, the number of viewpoints is assumed, and the number of pixels constituting one picture element Is set to k, the number of picture elements of each viewpoint video is obtained as horizontal k M / LX vertical N / k, and each viewpoint video is enlarged to horizontal MX vertical N and processed as necessary from each viewpoint video for each corresponding video area. A stereoscopic video processing method characterized by extracting and generating pixel data.

10. In the stereoscopic video processing method according to any one of claims 6 to 9, each viewpoint video is set to be larger by one pixel to several pixels on the left side, and the viewpoint image is generated on the left and right sides of the screen. A stereoscopic video processing method, wherein data extracted from the enlarged pixels is used in a data location.

11. The stereoscopic video processing method according to any one of claims 6 to 9, wherein black data is used in a non-data portion that occurs on the left and right of the screen. Method.

1 2. In the stereoscopic video processing method according to any one of claims 6 to 9, the non-data portion that is generated on the left and right of the screen is close to the non-data portion. A stereoscopic video processing method using copy data of a pixel at a viewpoint.

13. A stereoscopic video processing method according to claim 1 or 2, wherein a stereoscopic video image having a vertical parallax is generated.

1. A stereoscopic video display device comprising: a screen on which a video is displayed; and a separating unit for separating a position at which a pixel of each viewpoint video can be observed, the stereoscopic video display device according to any one of claims 1 to 13 When the image obtained by the image processing method is displayed on the screen, the aspect ratio of the screen pixel pitch is set such that the aspect ratio of the pitch of the displayed picture element group on the screen is 1: 1 to approximately 1: 1. A stereoscopic image display device.

15. The stereoscopic video display apparatus according to claim 14, wherein the number of viewpoints is k, and the number of pixels constituting one picture element is k. The pitch of display pixels is k (horizontal): L (vertical). Wherein the aspect ratio of the display picture element group is set such that width: length = 1: 1. 16. In a stereoscopic video display device equipped with a screen on which a video is displayed and a separating means for separating a position at which a pixel of each viewpoint video can be observed, the pitch of display pixels is k, the number of viewpoints is L, and the number of screen pixels is The aspect ratio of the pitch is set to k L to 1 to approximately k L to 1, and the image data is displayed by receiving the image data in which the pixel data of each viewpoint image is set sequentially in the horizontal direction and displayed on the screen A stereoscopic video display device characterized in that the aspect ratio of the pitch of the picture element group is 1: 1 to approximately 1: 1.

17. The stereoscopic image display device according to any one of claims 1 to 15, wherein the red pixel row, the color recording pixel row, and the blue pixel row are sequentially arranged in the vertical direction. A stereoscopic video display device characterized by the following.

18. The stereoscopic video display device according to claim 16, wherein a number of pixels of the same color corresponding to the number of viewpoint videos are continuously arranged. 3D image display device.